Cary Frydman, Peter Bossaerts, and Colin Camerer, California Institute of Technology; Nicholas C. Barberis, Yale University and NBER; and Antonio Rangel, California Institute of Technology and NBER
"Realization Utility and Regret Signals in the Brain are Associated with Suboptimal Stock Market Transactions"
    A growing body of evidence suggests that investors often make sub-optimal financial decisions. Some of these mistakes may result from value computations by the brain's decision making circuitry that induce individuals to maximize the hedonic experience that is associated with individual trades, instead of maximizing overall portfolio returns. Frydman, Bossaerts, Camerer, Barberis, and Rangel test this hypothesis by measuring neural activity with fMRI while subjects are making buying-and-selling decisions in an experimental stock market. With respect to selling mistakes, the researchers find that activity in the ventromedial prefrontal cortex and the ventral striatum (vSt) is correlated with a realization-utility signal at the time of making the selling decisions — the signal measures the profit or loss associated with selling individual stocks. The strength of the realization-utility signal is correlated with individuals' propensity to make selling mistakes. With respect to buying mistakes, it turns out that activity in the vSt is correlated with a regret signal at the time of price updates — that signal measures the forgone profits from not having purchased a successful stock. The strength of the regret signal is correlated with individuals' propensity to make buying mistakes.
    Discussant: Mark Dean, Brown University

Camelia M. Kuhnen, Northwestern University, and Brian Knutson and Gregory R. Samanez-Larkin, Stanford University
"Different Affective Learning Systems Contribute to the Accumulation of Assets and Debt"
    Neuroimaging findings suggest that individuals anticipate gains and losses with distinct neurophysiological systems. In a community sample, Kuhnen, Knutson, and Samanez–Larkin examined whether these systems might influence different types of learning (that is, gain versus loss) and contribute to different life financial outcomes (that is, assets versus debt). The researchers find that individuals who learned more rapidly about gains had more real-life assets, while individuals who learned more rapidly about losses had less real-life debt. Further, individuals whose medial prefrontal cortex best tracked gain-expected-value learned about gains more rapidly, while individuals whose anterior insula closely tracked loss-expected-value learned about losses more rapidly. These associations remained robust even after controlling for potential cognitive (for example working memory, cognitive flexibility, numeracy) and demographic (including age, sex, ethnicity, income, education) confounds. Beyond distinguishing systems that promote gain and loss learning, these findings imply that affective learning systems may promote different life financial outcomes.
    Discussant: Colin Camerer, Calltech

Henrik Cronqvist, Claremont McKenna College, and Stephan Siegel, Columbia University
"The Origins of Savings Behavior"
What are the origins of individual savings behavior? Using data on identical and fraternal twins matched with data on their savings behavior, Cronqvist and Siegel find that an individual's savings propensity is governed by both genetic predispositions, social transmission from parents to their children, and gene-environment interplay where certain environments moderate genetic influences. Genetic variation explains about 35 percent of the variation in savings rates across individuals, and this genetic effect is stronger in less constraining, high socioeconomic status environments. Parent-child transmission influences savings for young individuals and those who grew up in a family environment with less competition for parental resources. Individual-specific life experiences are a very important explanation for behavior in the savings domain, and strongest in urban communities. In a world progressing rapidly towards individual retirement savings autonomy, understanding the origins of individuals' savings behavior are of key importance to economists as well as policy makers.
    Discussant: Andrew Lo, MIT and NBER

Andrea Frazzini, AQR Capital Management, and Lasse H. Pedersen, New York University and NBER
"Betting Against Beta"
Frazzini and Pedersen present a model in which some investors are prohibited from using leverage and other investors' leverage is limited by margin requirements. The former group bid up high-beta assets while the latter group trades so as to profit from this, but must de-lever when they hit their margin constraints. The researchers test the model's predictions within U.S. equities, across 20 global equity markets, for Treasury bonds, corporate bonds, and futures. Consistent with their model, they find in each asset class that a betting-against-beta (BAB) factor — which is long a leveraged portfolio of low-beta assets and short a portfolio of high-beta assets -- produces significant risk-adjusted returns. When funding constraints tighten, betas are compressed towards one, and the return of the BAB factor is low.
    Discussant: John Heaton, University of Chicago NBER

Xavier Gabaix, New York University and NBER
"A Sparsity-Based Model of Bounded Rationality"
    Gabaix proposes and analyzes a model with boundedly rational features in which the decisionmaker (DM) behaves like an economist who builds a simplified representation of the world. Crucially, this representation uses few parameters that are non-zero, or that differ from the usual state of affairs. The DM may not maximize perfectly , again based on a penalty related to sparsity — the lack of sparsity is formulated so as to lead to well-behaved, convex maximization problems. Gabaix applies the model to a variety of prototypical economic situations: hitting a target with selective attention; picking a consumption bundle, but with imperfect understanding of price; optimal pricing with boundedly rational consumers which, when paired with optimal response by firms, generates a novel mechanism for price rigidity; life-cycle consumption and investment problems; failures of Euler equations; portfolio choice problems with stocks and flows; the aquiring-a-company problem; and the dollar auction game. He concludes that the model may represent a useful proposal for tractable analysis of bounded rationality in economic situations.
    Discussant: Sendhil Mullainathan, Harvard University and NBER

Conference sponsored by
Bracebridge Capital, Fuller & Thaler Asset Management and
LSV Asset Management
under the auspices of the National Bureau of Economic Research